Periodic Reporting for period 1 - AM2PM (ADDITIVE TO PREDICTIVE MANUFACTURING FOR MULTISTOREY CONSTRUCTION USING LEARNING BY PRINTING AND NETWORKED ROBOTICS)
Período documentado: 2024-10-01 hasta 2025-09-30
Context and Rationale
The construction sector is one of the largest contributors to climate change, responsible for nearly 40% of global energy use and CO2 emissions, with cement accounting for 8%. Cement demand is expected to rise by up to 23% by 2050, while construction and demolition waste already represents about 40% of global solid waste. Without intervention, these trends threaten Europe’s ability to meet EU Green Deal and Fit for 55 targets. Yet this challenge also offers an opportunity. 3D concrete printing (3DCP) provides a radically different construction model, with potential reductions of 60% in material use, 80% in cost, and 70% in construction time. However, adoption remains limited due to gaps in material design, process control, scalability, and environmental integration. A shortage of skilled labour further complicates deployment, heightening the need for integrated robotic fabrication workflows. Current approaches remain fragmented, heavily reliant on trial-and-error and open-loop processes unsuited for industrial-scale impact.
Project Objectives
AM2PM combines material, digital, and robotic innovation to create an AI-driven Digital Twin platform uniting digital manufacturing and robotic construction. Laboratory demonstrators and pilot studies will validate this ecosystem, enabling multi-storey cyber-physical construction workflows. The project aims to set new standards, methods, and materials to accelerate decarbonisation and digitalisation of the AEC sector.
Central to this effort is the development of sustainable cementitious materials using recycled granular waste and locally sourced resources, targeting a 50% reduction in embodied carbon and significantly reduced cement use. In parallel, AM2PM advances computational structural design by integrating buildability, material efficiency, and environmental metrics directly into topology optimisation, generating complex load-bearing components that are both robust and resource-efficient.
To ensure reliability at scale, the project introduces Learning-by-Printing (LbP), an AI-driven methodology for real-time prediction, monitoring, and correction of 3DCP processes. These adaptive models integrate with dynamic Digital Twin systems that coordinate robots, sensors, and construction workflows through continuous bi-directional feedback. This ensures accuracy, adaptability, and resilience in real construction environments. Predictive Life Cycle Assessment (LCA) is embedded from the earliest design stages, enabling environmental impacts to be anticipated and mitigated before fabrication and construction. Together, these innovations define a digital, resource-efficient, and climate-conscious framework for next-generation construction.
Pathway to Impact
AM2PM’s pathway to impact merges technical breakthroughs with system-level transformation across materials science, digital design, robotics, AI, and multi-scale integration. Laboratory demonstrators and pilots will lead to a cyber-physical platform for multistorey building components. Beyond technological achievements, AM2PM will shape new standards and workflows for the AEC sector, embedding predictive LCA, automation, and scalable digital practices into mainstream construction. Aligned with the European Green Deal and the New European Bauhaus, the project couples sustainability with digital and cultural innovation, supporting Europe’s leadership in low-carbon construction technologies.
The construction sector is one of the largest contributors to climate change, responsible for nearly 40% of global energy use and CO2 emissions, with cement accounting for 8%. Cement demand is expected to rise by up to 23% by 2050, while construction and demolition waste already represents about 40% of global solid waste. Without intervention, these trends threaten Europe’s ability to meet EU Green Deal and Fit for 55 targets. Yet this challenge also offers an opportunity. 3D concrete printing (3DCP) provides a radically different construction model, with potential reductions of 60% in material use, 80% in cost, and 70% in construction time. However, adoption remains limited due to gaps in material design, process control, scalability, and environmental integration. A shortage of skilled labour further complicates deployment, heightening the need for integrated robotic fabrication workflows. Current approaches remain fragmented, heavily reliant on trial-and-error and open-loop processes unsuited for industrial-scale impact.
Project Objectives
AM2PM combines material, digital, and robotic innovation to create an AI-driven Digital Twin platform uniting digital manufacturing and robotic construction. Laboratory demonstrators and pilot studies will validate this ecosystem, enabling multi-storey cyber-physical construction workflows. The project aims to set new standards, methods, and materials to accelerate decarbonisation and digitalisation of the AEC sector.
Central to this effort is the development of sustainable cementitious materials using recycled granular waste and locally sourced resources, targeting a 50% reduction in embodied carbon and significantly reduced cement use. In parallel, AM2PM advances computational structural design by integrating buildability, material efficiency, and environmental metrics directly into topology optimisation, generating complex load-bearing components that are both robust and resource-efficient.
To ensure reliability at scale, the project introduces Learning-by-Printing (LbP), an AI-driven methodology for real-time prediction, monitoring, and correction of 3DCP processes. These adaptive models integrate with dynamic Digital Twin systems that coordinate robots, sensors, and construction workflows through continuous bi-directional feedback. This ensures accuracy, adaptability, and resilience in real construction environments. Predictive Life Cycle Assessment (LCA) is embedded from the earliest design stages, enabling environmental impacts to be anticipated and mitigated before fabrication and construction. Together, these innovations define a digital, resource-efficient, and climate-conscious framework for next-generation construction.
Pathway to Impact
AM2PM’s pathway to impact merges technical breakthroughs with system-level transformation across materials science, digital design, robotics, AI, and multi-scale integration. Laboratory demonstrators and pilots will lead to a cyber-physical platform for multistorey building components. Beyond technological achievements, AM2PM will shape new standards and workflows for the AEC sector, embedding predictive LCA, automation, and scalable digital practices into mainstream construction. Aligned with the European Green Deal and the New European Bauhaus, the project couples sustainability with digital and cultural innovation, supporting Europe’s leadership in low-carbon construction technologies.
AM2PM achieved major advances over the past year. In materials research, the project validated mixes incorporating marginal, waste, and recycled aggregates using new in-situ testing protocols (WP1), supported by WP5 pilot studies. A computational framework was developed for prestressed concrete that links optimisation of printed formwork, casting, and prestressing with robotic toolpaths and buildability constraints (WP2). A cyber-physical infrastructure was established combining robotics, vision systems, sensors, and AI into a Learning-by-Printing feedback loop, achieving predictive quality control validated in multiple experiments (WP3). At the building-system scale, hackathons and stakeholder workshops helped define high-rise construction configurations, producing sequences, component catalogues, and human-robot collaboration frameworks (WP4). AM2PM also conducted requirements analysis and defined the methodological basis for a predictive LCA framework linking component-level assessments with whole-building evaluations aligned with regulations, now being integrated into the Digital Twin (WP6). The project strengthened its contribution to the EIC Pathfinder ecosystem through DigiTrio portfolio activities, including strategic planning, participation in working groups, and the development of a unified brand identity (WP7). Dissemination and exploitation activities were consolidated through online platforms, newsletters, communication materials, and the identification of exploitable results with market and IPR strategies (WP8). Coordination has been effective through annual meetings, monthly online reviews, collaboration with the Project Officer, and integration across WPs (WP9).
AM2PM demonstrated that conventional cast-specimen tests and simple buildability metrics fail to predict long-term performance of printed materials, confirming the need for in-situ protocols capturing early viscoelastic and surface behaviour (WP1). The project delivered one of the most advanced computational frameworks for 3DCP, uniting optimisation, material parameters, and robotic fabrication into a single workflow (WP2). The Learning-by-Printing methodology achieved more than 95% accuracy, enabling adaptive, reproducible process control across scales and materials (WP3). A comprehensive System-of-Systems analysis was established for multi-storey robotic construction, integrating machines, materials, workflows, and human-robot interaction (WP4). The project introduced a methodological LCA framework for 3DCP with unprecedented granularity and comparability (WP6). Portfolio-level communication was unified (WP7), dissemination exceeded expectations (WP8), and coordination accelerated cross-WP integration (WP9).